Wake, sleep and circadian disturbances are common occurrences in Alzheimer' disease (AD), often times preceding amnestic symptoms. Such disturbances affect the quality of life of patients and caregivers alike and boost institutionalization. A bidirectional correlation between amyloid-beta (A?) deposition and disturbed sleep contributes to slow wave sleep (SWS) deficits and sleep fragmentation. We discovered converging evidence in human sleep and neuropathological studies suggesting that individuals with progressive supranuclear palsy (PSP), a primary tauopathy, show an extreme sleep phenotype featuring a much shorter sleep duration. This point for a role of tau-related degeneration as an underlying cause of sleep disfunction, independent of A? deposition. Interestingly, brainstem, hypothalamic and basal forebrain nuclei involved in circandian-sleep-wake regulation develop AD- tau-based neurofibrillary tangles preceding tangles in cortical areas and often, before A? plaques appear. Our working hypothesis is that tau-induced degeneration of key brainstem, hypothalamic and basal forebrain nuclei controlling 1) SWS; 2) waking-arousal; and 3) circadian timing underlie sleep-wake behavior in AD, preceding both cognitive decline and later emergence of the feedforward cycle of sleep disturbance and accelerated A? deposition. We will test our hypothesis contrasting sleep-wake behavior in progressive AD stages versus healthy controls by analyzing differences in objective sleep measurements, clinical and molecular imaging profiles and quantitative pathoanatomical measures in nuclei involved in wake, NREM sleep regulation and circadian rhythm. Moreover, we will add a PSP as a positive control group. We are uniquely poised to succeed due to our group expertise, track record of working together and our access to uniquely well characterized clinicopathological cohort. This combination of factors creates a unique opportunity to exploit novel human findings that will inform and complement mechanistic hypotheses and testing in model systems. This is critical because animals' sleep-wake patterns and AD-like models diverge from those of humans and experimental models rather mimic non-AD tauopathies than tau-related AD patterns. We anticipate our findings will inform critical information on the temporal sequence of disrupted sleep and/or circadian rhythms and the accumulation and spreading of protein aggregates such as phospho-tau and A? in AD. Beyond this, results from this study will inform rational therapies for treating disturbed sleep in AD.